Dazhuo Qiu

Explaining the behavior of graph neural networks (GNNs) has become critical due to their "black-box'' nature, especially in the context of analytical tasks such as graph classification. Current approaches are limited to providing explanations for individual instances or specific class labels and may return large explanation structures that are hard to access, nor directly queryable. In this paper, we present GVEX [1] (Graph Views for GNN EXplanation) -- our system developed to offer user-friendly, interactive, and configurable explanations for GNNs based on graph views. GVEX provides a configuration component to enable users to easily select a desired number of important nodes from different classes, thereby generating explanations tailored to multiple classes of interest. Furthermore, GVEX generates high-quality explanation subgraphs by identifying important nodes exploiting factual and counterfactual properties and by computing their aggregated influence on the remaining nodes following the GNN message passing paradigm. Lastly, GVEX performs a summarize step on top of lower-tier explanation structures to generate higher-tier graph patterns that offer direct access for users with (domain-aware) queries. Our demonstration will highlight (1) a novel two-tier explanation structure called explanation views, consisting of graph patterns and a set of explanation subgraphs, which provide high-quality explanations for GNNs; (2) the system's intuitive GUI facilitates user interaction to configure personalized settings, e.g., classes of interest and explanation size, and compare with other explanation algorithms; (3) GVEX generates queryable explanations, making it easy for human experts to access and inspect with domain knowledge. Our demonstration video is at: https://youtu.be/q9d7ldulIuQ.

Generating explanations for graph neural networks (GNNs) has been studied to understand their behaviors in analytical tasks such as graph classification. Existing approaches aim to understand the overall results of GNNs rather than providing explanations for specific class labels of interest, and may return explanation structures that are hard to access, nor directly queryable. We propose GVEX, a novel paradigm that generates Graph Views for GNN EXplanation. (1) We design a two-tier explanation structure called explanation views. An explanation view consists of a set of graph patterns and a set of induced explanation subgraphs. Given a database G of multiple graphs and a specific class label l assigned by a GNN-based classifier M, it concisely describes the fraction of G that best explains why l is assigned by M. (2) We propose quality measures and formulate an optimization problem to compute optimal explanation views for GNN explanation. We show that the problem is Σ2P-hard. (3) We present two algorithms. The first one follows an explain-and-summarize strategy that first generates high-quality explanation subgraphs which best explain GNNs in terms of feature influence maximization, and then performs a summarization step to generate patterns. We show that this strategy provides an approximation ratio of 1/2. Our second algorithm performs a single-pass to an input node stream in batches to incrementally maintain explanation views, having an anytime quality guarantee of 1/4-approximation. Using real-world benchmark data, we experimentally demonstrate the effectiveness, efficiency, and scalability of GVEX. Through case studies, we showcase the practical applications of GVEX.

With the proliferation of wireless and mobile devices, Spatial Crowdsourcing (SC) attracts increasing attention, where task assignment plays a critically important role. However, recent task assignment solutions in SC often assume that data is stored in a central station while ignoring the issue of privacy leakage. To enable decentralized training and privacy protection, we propose a federated task assignment framework with personalized location-preference learning, which performs efficient task assignment while keeping the data decentralized and private in each platform center (e.g., a delivery center of an SC company). The framework consists of two phases: personalized federated location-preference learning and task assignment. Specifically, in the first phase, we design a personalized location-preference learning model for each platform center by simultaneously considering the location information and data heterogeneity across platform centers. Based on workers' location preference, the task assignment phase aims to achieve effective and efficient task assignment by means of the Kuhn-Munkres (KM) algorithm and the newly proposed conditional degree-reduction algorithm. Extensive experiments on real-world data show the effectiveness of the proposed framework.